Cognitive Scaffold

Preparing your thinking workspace

arrow_back_ios_new
MENTAL MODEL · M1755

Overfitting

Overfitting
TechnicalHigh supportData Science
Included
account_tree

Version 1.0.0 · Updated 2026-07-30

CORE DEFINITION

A model performs too well on training data (memorizing every noise point), losing its generalization ability and performing poorly on new data. Because it learns 'special cases' rather than 'patterns'.

SCAFFOLDING EFFECT

psychology

Reduce cognitive load

- Empiricism trap: Do not treat the specific conditions of a one-time success (e.g., buying a house in 2010 to get rich) as a universal rule. That might just be 'overfitting' to the market of that year. When facing new environments, retain a certain degree of 'fuzziness' and adaptability.

anchor

Anchor fast decisions

The model learns noise as patterns too rigidly on the training set, causing a sharp drop in performance on new data. The core tension is between 'fitting' and 'generalization'.

MINIMUM ACTION

In progress 0/4

Practice this model in one real situation:

Check to track your progress (stored locally)
Learning progress0%
account_treeGenealogyexpand_more
menu_bookReferencesexpand_more

Source support: Explicit

  • link
    en.wikipedia.orghttps://en.wikipedia.org/wiki/OverfittingZH · Explicit
    verified

RELATED MODELS

Related model

Datasaurus Dozen

A set of datasets with nearly identical summary statistics whose scatterplots look completely different, proving that numbers alone can mislead.

arrow_forwardRelated model

Bias-Variance Tradeoff

The tension in model building between bias, which leaves a model too rigid to capture real structure, and variance, which leaves it flexible enough to fit noise.

arrow_forwardRelated model

Double Descent

A learning phenomenon in which test error falls, rises, then falls again as model complexity grows past the interpolation threshold.

arrow_forward